apextoaster
/
onnx-web
1
0
Fork 0
Code Actions

restore original LPW names

This commit is contained in:
Sean Sube 2023-02-18 18:54:24 -06:00
parent 4d93c13431
commit 2b83f942af
Signed by: ssube
GPG Key ID: 3EED7B957D362AF1
2 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
api/onnx_web/chain/blend_inpaint.py
View File

@ -85,7 +85,7 @@ def blend_inpaint(
height=size.height, height=size.height,
image=tile_source, image=tile_source,
latents=latents, latents=latents,
mask=tile_mask, mask_image=tile_mask,
negative_prompt=params.negative_prompt, negative_prompt=params.negative_prompt,
num_inference_steps=params.steps, num_inference_steps=params.steps,
width=size.width, width=size.width,
@ -100,7 +100,7 @@ def blend_inpaint(
height=size.height, height=size.height,
image=tile_source, image=tile_source,
latents=latents, latents=latents,
mask=mask, mask_image=mask,
negative_prompt=params.negative_prompt, negative_prompt=params.negative_prompt,
num_inference_steps=params.steps, num_inference_steps=params.steps,
width=size.width, width=size.width,

22
api/onnx_web/diffusion/lpw_stable_diffusion_onnx.py
View File

@ -657,7 +657,7 @@ class OnnxStableDiffusionLongPromptWeightingPipeline(OnnxStableDiffusionPipeline
prompt: Union[str, List[str]], prompt: Union[str, List[str]],
negative_prompt: Optional[Union[str, List[str]]] = None, negative_prompt: Optional[Union[str, List[str]]] = None,
image: Union[np.ndarray, PIL.Image.Image] = None, image: Union[np.ndarray, PIL.Image.Image] = None,
mask: Union[np.ndarray, PIL.Image.Image] = None, mask_image: Union[np.ndarray, PIL.Image.Image] = None,
height: int = 512, height: int = 512,
width: int = 512, width: int = 512,
num_inference_steps: int = 50, num_inference_steps: int = 50,
@ -687,9 +687,9 @@ class OnnxStableDiffusionLongPromptWeightingPipeline(OnnxStableDiffusionPipeline
image (`np.ndarray` or `PIL.Image.Image`): image (`np.ndarray` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, that will be used as the starting point for the `Image`, or tensor representing an image batch, that will be used as the starting point for the
process. process.
mask (`np.ndarray` or `PIL.Image.Image`): mask_image (`np.ndarray` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be `Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be
replaced by noise and therefore repainted, while black pixels will be preserved. If `mask` is a replaced by noise and therefore repainted, while black pixels will be preserved. If `mask_image` is a
PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should
contain one color channel (L) instead of 3, so the expected shape would be `(B, H, W, 1)`. contain one color channel (L) instead of 3, so the expected shape would be `(B, H, W, 1)`.
height (`int`, *optional*, defaults to 512): height (`int`, *optional*, defaults to 512):
@ -782,10 +782,10 @@ class OnnxStableDiffusionLongPromptWeightingPipeline(OnnxStableDiffusionPipeline
image = preprocess_image(image) image = preprocess_image(image)
if image is not None: if image is not None:
image = image.astype(dtype) image = image.astype(dtype)
if isinstance(mask, PIL.Image.Image): if isinstance(mask_image, PIL.Image.Image):
mask = preprocess_mask(mask, self.vae_scale_factor) mask_image = preprocess_mask(mask_image, self.vae_scale_factor)
if mask is not None: if mask_image is not None:
mask = mask.astype(dtype) mask = mask_image.astype(dtype)
mask = np.concatenate([mask] * batch_size * num_images_per_prompt) mask = np.concatenate([mask] * batch_size * num_images_per_prompt)
else: else:
mask = None mask = None
@ -1057,7 +1057,7 @@ class OnnxStableDiffusionLongPromptWeightingPipeline(OnnxStableDiffusionPipeline
def inpaint( def inpaint(
self, self,
image: Union[np.ndarray, PIL.Image.Image], image: Union[np.ndarray, PIL.Image.Image],
mask: Union[np.ndarray, PIL.Image.Image], mask_image: Union[np.ndarray, PIL.Image.Image],
prompt: Union[str, List[str]], prompt: Union[str, List[str]],
negative_prompt: Optional[Union[str, List[str]]] = None, negative_prompt: Optional[Union[str, List[str]]] = None,
strength: float = 0.8, strength: float = 0.8,
@ -1079,9 +1079,9 @@ class OnnxStableDiffusionLongPromptWeightingPipeline(OnnxStableDiffusionPipeline
image (`np.ndarray` or `PIL.Image.Image`): image (`np.ndarray` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, that will be used as the starting point for the `Image`, or tensor representing an image batch, that will be used as the starting point for the
process. This is the image whose masked region will be inpainted. process. This is the image whose masked region will be inpainted.
mask (`np.ndarray` or `PIL.Image.Image`): mask_image (`np.ndarray` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be `Image`, or tensor representing an image batch, to mask `image`. White pixels in the mask will be
replaced by noise and therefore repainted, while black pixels will be preserved. If `mask` is a replaced by noise and therefore repainted, while black pixels will be preserved. If `mask_image` is a
PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should PIL image, it will be converted to a single channel (luminance) before use. If it's a tensor, it should
contain one color channel (L) instead of 3, so the expected shape would be `(B, H, W, 1)`. contain one color channel (L) instead of 3, so the expected shape would be `(B, H, W, 1)`.
prompt (`str` or `List[str]`): prompt (`str` or `List[str]`):
@ -1136,7 +1136,7 @@ class OnnxStableDiffusionLongPromptWeightingPipeline(OnnxStableDiffusionPipeline
prompt=prompt, prompt=prompt,
negative_prompt=negative_prompt, negative_prompt=negative_prompt,
image=image, image=image,
mask=mask, mask_image=mask_image,
num_inference_steps=num_inference_steps, num_inference_steps=num_inference_steps,
guidance_scale=guidance_scale, guidance_scale=guidance_scale,
strength=strength, strength=strength,